GNU Arch - Definition and Overview

In computing, GNU arch is a revision control system, similar in purpose to tools such as CVS, SCCS, and Subversion. It is used to keep track of the changes made to a source tree and to help programmers combine and otherwise manipulate changes made by multiple people or at different times.

GNU arch uses a slightly different paradigm than most versioning systems, in that each revision is uniquely globally identifiable. This results in a very scalable system that allows easy merging and applying of changes from completely disparate sources.

GNU arch is decentralized, removing the need for a central server for which developers have to be authorized in order to contribute. Instead, GNU arch is designed so that a full read-only copy of a project is made accessible by a head developer via HTTP, FTP, or SFTP, and each contributor is encouraged to retrieve a copy of the project, make modifications, then publish their changeset to allow the head developer to manually merge said changeset into the official project that's later refreshed on the read-only copy.

To simulate the behavior of centralized revision control systems, the head developer could allow shell access (for example, SSH) to the server hosting the aforementioned read-only copy. Authorized users would have write access via the filesystem.

Some other features of GNU arch include:

• atomic commits: the commit is all-or-nothing—if it is interrupted before it is fully complete, the entire commit is rolled back until better times.
• changeset orientation: instead of tracking individual files, GNU arch tracks changesets (individual features or bugfixes), which may span multiple files.
• easy and sophisticated branching
• advanced merging
• cryptographic signatures to ensure integrity

External links

• The GNU Arch homepage (http://www.gnu.org/software/gnu-arch/)
• The arch wiki (http://wiki.gnuarch.org)
• An extensive tutorial on arch (http://regexps.srparish.net/tutorial-tla/arch.html)